chengping shen shencp@phys.hawaii

Chengping Shenshencp@phys.hawaii.edu

Phys. Dept., University of Hawaii

BEPC II Storage ringBEPC II Storage ring:: Large angle, double-ring

RFRF SR

IP

22 m

rad

2. 5m8ns

1. 5cm

0.1cm

Beam energy: 1.0-2.3 GeVLuminosity: 1×1033 cm-2s-1

Optimum energy: 1.89 GeVEnergy spread: 5.16 ×10-4

No. of bunches: 93Bunch length: 1.5 cmTotal current: 0.91 ASR mode: 0.25A @ 2.5 GeV

• Drift chamber and its electronics (IHEP, Sichuan, Tsinghua)

• CsI(Tl) calorimeter and its electronics(IHEP, Tsinghua)• TOF （ IHEP, USTC ， Tokyo ， Hawaii ）• TOF electronics （ USTC ）• RPC （ IHEP, Uni. of Washington ）• RPC electronics （ USTC ）• Trigger （ IHEP, USTC ）• DAQ & online software (IHEP, Tsinghua)• Offline software （ IHEP, Peking ， Shangdong ， Nanji

ng ）• Superconducting magnet （ IHEP ， Wang NMR ）• Mechanics (IHEP)• Technical support ( IHEP, Tsinghua ）

Physics at BEPCII/BESIII• Precision measurement of CKM matrix elements• Precision test of Standard Model• QCD and hadron production• Light hadron spectroscopy• Charmonium physics• Search for new physics/new particles

Physics

Channel

Energy

(GeV)

Luminosity

(1033 cm–2s –1)

Events/year

J/ 3.097 0.6 1.0×1010

3.67 1.0 1.2×107

’ 3.686 1.0 3.0 ×109

D* 3.77 1.0 2.5×107

Ds 4.03 0.6 1.0×106

Ds 4.14 0.6 2.0×106

A review book on tau-charm physics at BESIII~ 800 pages, to be completed this year

BOSS--BESIII Offline Software System• There are two ways to install BOSS

1. Using “pacman” method: easy to install, save time. But if you have sth. wrong in stalling, you must remove everything, and start again. You also need to install database by yourself.

2. Using regular method: we have written a simple manual based BOSS6.3.0. The BOSS6.3.1, the method should be similar. But this method needs much time, and you need to copy and correct many files by yourself.

Use pacman to install

Boss development environment(BOSS6.3.1)

Machine: lxslc01(-03).ihep.ac.cn(SLC3), lxslc04(-06) ).ihep.ac.cn(SLC4)

OS : Scientific Linux Cern

Compiler : gcc-3.4.6

Framework :Based on GAUDI-v1r14p20031120

External Libs: CERNLIB 、 ROOT 、 CLHEP 、 Geant4 、 and etc.

Developing language : C++ ， some FORTRAN

Database : PostgreSql, MySql

Version management tool : CVS

Browsing files in CVS repository : ViewCVS

Configuration management tool : CMT

BOSS Structure

Analysis -- Analysis software and some analysis tools

BesExamples -- Examples for using some packages

BesPolicy -- define some standard patterns of Bes

BesCxxPolicy -- define some standard patterns only for c++

compilier

BesFortranPolicy -- define some standard patterns only for

fortran compilier

BesRelease -- for all the packages release

Calibration -- for calibration framework and common tools

Control -- Common tools and framework of boss

Database -- the geometry data IO

DetectorDescription -- Detector Description packages Emc -- for emc Event -- different event model EventDisplay -- the interface for event display EventFilter -- for event filter External -- interfaces of external libs Generator -- for generators MagneticField -- Mdc -- for mdc

Muc -- for MuonCounter

Reconstruction -- all reconsrtuction algorithms

RootPolicy --

Simulation -- simulation

TestRelease -- for some test release

Tof -- tof

Utilities -- some common tools

CVS and CMT

The BESIII Offline Software sources are stored in CVS and can be accessed using the CVS command.

The BESIII Offline Software libraries have been built using the CMT. Therefore, using the CMT tool is the recommended way to modify existing packages or re-build the examples included in the release.

http://docbes3.ihep.ac.cn/viewvc/cgi-bin/viewvc.cgi/BossCvs/

What is CMT （ Configuration management tool ）

• A set of tools and conventions– structures software development or production

• concepts of areas, packages, versions, constituents

– organises software into packages– describes package properties– describes package constituents– operates the software production (management, build,

import/export, etc...)• by transparently configuring and driving the various

conventional tools (CVS, make, MSDev, Web, tar, compilers, linkers, archivers, etc...)

Useful variables and commands

• Basic syntaxcmt <verb> [ -<option>=<value> <parameters…> ]

• Main commands cmt config re-configure a package cmt create … create a new package cmt show … monitors configuration parameters cmt broadcast … launch shell command over several packages

cmt co … interface to CVS’ checkout operation cmt help get the list of possible commands source cleanup.sh undo the source setup source setup.sh do the source setup

Structure of a package 　

• /src source files (*.cxx)• /<package-name> header files (*.h)• /cmt requirements file and some scripts

• /share jobOptions and other

running scripts• /i386_linux24 Automatically created at build time

(libs. Applications and etc. under this directory)

Creating a new package

1 、 Choosing a location

2 、 Choosing a name– Must be a unique package name in boss

(should not either just differentiate by case of letters)

3 、 Choosing a version

4 、 The command

cmt create BesGeoDB BesGeoDB-00-00-01

Examples…• The simplest one

– One new application with (some) source file(s)

>cmt create A A-00-00-01>cd A/A-00-00-01/cmt>vi requirements>vi A.cxx >gmake>source setup.csh>cd ../i386-linux24>./A.exe

package Ause BesPolicy BesPolicy-01-*application A A.cxx

1) Create a new packageand move to it

2) Describe the package

3) Create the package items

and build them

4) set the environment

and run

The requirements file

The requirements file

The requirements file centralizes all configuration parametersAuthor(s),

manager(s)

Structural information •specialized directory structure •used packages•links to external packages

Constituents•Libraries•Applications•generated documents

Make macros and environment variables

and their possible values on various platforms, sites, environments Customization for

new languages, or document generators

Definition of conventional behavioural

patterns

Sufficient for transparently generating all required makefiles and setup scripts, or for

launching test procedures. They are the sources of all queries.

Example of requirements(package)package getacAlgauthor shencp@ihep.ac.cn

use BesPolicy BesPolicy-01-*use GaudiInterface GaudiInterface-01-* Externaluse DstEvent DstEvent-* Eventuse EventModel EventModel-* Eventuse MdcRawEvent MdcRawEvent-* Mdcuse ParticleID ParticleID-* Analysisuse VertexFit VertexFit-* Analysisuse BesROOT BesROOT-00-* Externaluse getacAlg TemplateAlg-* Analysis/Physicsuse McTruth McTruth-* Eventuse AnalEvent AnalEvent-* Event

apply_pattern component_librarylibrary getacAlg *.cxx components/*.cxx

Checking out a package• Getting a package out from CVS to

modify it– Go to your private development area– Check out the package (installing the proper

CMT directory structure) cvs login

cmt co TestRelease (or) cmt co –r TestRelease-00-00-15 TestRelease

– Get the required revision, and make use of that tag--Applying a command to all checked out packages

cmt broadcast cmt config cmt broadcast gmake

Running boss

#running a simple example -- ExHelloWorld• cd <yourworkarea>• cmt co TestRelease• cd Testrelease/*/cmt• jed requirements Add “use ExHelloWorld ExHelloWorld-* BesExamples” to requirements

• cmt broadcast cmt config• cmt broadcast gmake• source setup.sh• cd ../run• source ../share/TestRelease_links.csh• boss.exe HelloWorldOptions.txt

#include "$KKMCROOT/share/jobOptions_KKMC.txt“

KKMC.CMSEnergy = 3.097;KKMC.CMSEnergySpread=0.0008;KKMC.InitializedSeed={400206,1,0};KKMC.NumberOfEventPrinted=100;KKMC.GenerateJPsi=true;

EvtDecay.userDecayTableName = "jpsi.dec";

KKMC.ParticleDecayThroughEvtGen=true;

BESIII used KKMC+BesEvtGen togenerate charmonium decay events.The KKMC is used to simulate e+e-annihilation till ccbar production including ISR effects, together with theBeam energy spread.

Set CMS energy, energy spread, Initialized seed and the number ofprinted events(output topology)

User define decay table, here jpsi.dec define all the decays of jpsi (inclusive sample)

ApplicationMgr.DLLs += { "BesEvtGen","BesServices","BesSim"};ApplicationMgr.TopAlg += { "EvtDecay","BesSim"};

#include "$BESSIMROOT/share/Bes_Gen.txt"#include "$BESSIMROOT/share/G4Svc_BesSim.txt"

#include "$CALIBSVCROOT/share/job-CalibData.txt"#include "$MDCCALIBFUNSVCROOT/share/job_MdcCalibFunSvc.txt"CalibRootCnvSvc.Mdcrootfile="/afs/ihep.ac.cn/bes3/offline/sw/data/constant/MdcCalibConst.root";

G4Svc.RunID = -10206;

G4Svc.FADSMacro = "run.mac";// collision with beam angle G4Svc.BoostLab = true;

Run Number

Geant4 run macro file, control MDC noise

Define the angle between the e- beam and e+ beam

// beam bunch position, unit (mm)G4Svc.BeamPosX=0;G4Svc.BeamPosY=0;G4Svc.BeamPosZ=0;

// beam bunch size , unit (mm)G4Svc.BeamSizeX=0.38;G4Svc.BeamSizeY=0.0057;G4Svc.BeamSizeZ=15;

// geometry construction(0:not construct, 1:construct with G4 code)// (2:construct with gdml files)BesSim.Mdc = 2;BesSim.Tof = 2;BesSim.Emc = 2;BesSim.Muc = 2;BesSim.PipeSCM = 2;

#include "Sim2Root_Options.txt"RootCnvSvc.digiRootOutputFile = "./mc10206-2.rtraw";

// OUTPUT PRINTOUT LEVEL// Set output level threshold (2=DEBUG, 3=INFO, 4=WARNING, 5=ERROR, 6=FATAL )MessageSvc.OutputLevel = 6;

ApplicationMgr.EvtMax = 500; Total Number need to be produced

Output file name, not ASCII file any more. (new to BOSS6.3.1)

Generator

The framework of BESIII event generator is approved to use

KKMC+BesEvtGen to generate events.

e

e

' '

J /

(2S)

X

*

' '

e e (n )

(n )J /

(n )

J / X

(2S) X

X

KKMC(ISR) + BesEvtGen(FSR)

Both for exclusive and inclusive decays

BESIII Generator Framework

GenBes: obsolete

Not recommend to use

CMSLab. Syst. (Yan Ye)

G4

#include "ReadRoot_Options.txt"#include "$CALIBSVCROOT/share/job-CalibData.txt"#include "$ESTIMEALGROOT/share/job_EsTimeAlg.txt"//zhangy#include "$MDCTRKRECONROOT/share/jobOptions_MdcPatRec.txt" //zangsl//#include "$TRKRECOROOT/share/jobOptions_TrkReco.txt"

#include "$KALFITALGROOT/share/job_kalfit_all.txt" #include "$MDCDEDXALGROOT/share/job_dedx_all.txt"#include "$TRKEXTALGROOT/share/TrkExtAlgOption.txt"#include "$EMCRECROOT/share/EmcRecOptions.txt"#include "$TOFRECROOT/share/jobOptions_TofRec.txt"#include "$MUCRECALGROOT/share/jobOptions_MucRecTrkExt.txt"#include "$RECEVENTASSEMBLYALGROOT/share/RecEventAssemblyAlg_Options.txt"

//in order to save MC Truth information#include "Save2Root_Options.txt"

36

// Set output level threshold (2=DEBUG, 3=INFO, 4=WARNING, 5=ERROR, 6=FATAL )ApplicationMgr.DLLs += {"EmcRecGeoSvc","RootHistCnv","DedxCorrecSvc", "TofGeomSvc","TofCaliSvc","MucGeomSvc" }; MessageSvc.OutputLevel = 5; // Number of events to be processed (default is 10) ApplicationMgr.EvtMax = 50000; EventCnvSvc.digiRootInputFile = {"./jpsi_getac_etapipi.rtraw" }; ApplicationMgr.HistogramPersistency = "ROOT"; EventCnvSvc.digiRootOutputFile ="./jpsi_getac_etapipi.rec";

NTupleSvc.Output = { "FILE101 DATAFILE='/ihepbatch/d03/shencp/tmp/Mdc.root' OPT='NEW' TYP='ROOT'", "FILE102 DATAFILE='/ihepbatch/d03/shencp/tmp/fzisan.root' OPT='NEW' TYP='ROOT'", "FILE104 DATAFILE='/ihepbatch/d03/shencp/tmp/kalfit.root' OPT='NEW' TYP='ROOT'", "FILE103 DATAFILE='/ihepbatch/d03/shencp/tmp/DeDx.root' OPT='NEW' TYP='ROOT'", "FILE105 DATAFILE='/ihepbatch/d03/shencp/tmp/Estime.root' OPT='NEW' TYP='ROOT'", "FILE201 DATAFILE='/ihepbatch/d03/shencp/tmp/Tof.root' OPT='NEW' TYP='ROOT'", "FILE301 DATAFILE='/ihepbatch/d03/shencp/tmp/Emc.root' OPT='NEW' TYP='ROOT'", "FILE401 DATAFILE='/ihepbatch/d03/shencp/tmp/Muc.root' OPT='NEW' TYP='ROOT'",

"FILE998 DATAFILE='/ihepbatch/d03/shencp/tmp/RecAssembly.root' OPT='NEW' TYP='ROOT'"};

(In your analysis job option)

Another method on PIDThe ANN method is contained in the Analysis/PsrticleID package. To use the ANN method, these codes are suggested: pid->init();     pid->setDstTrack(*itTrk);       pid->usePidSys(pid->useDedx() | pid->useTof1() | pid->useTof2() | pid->useEmc() | pid->useMuc());       pid->identify(pid->all()|pid->onlyMuon());       pid->setNeuronPidCuts();       pid->setMethod(pid->methodNeuronNetwork());     pid->calculate();    if(!(pid->neuronStat()> 0)) continue;        // the detectors' information is not correct if (pid->neuronStat()< 0)        if( pid->neuronVal()>0.5 &&  pid->neuronVal()<1.5 )         //the particle is electron if 0.5<pid->neuronVal()<1.5    if( pid->neuronVal()>1.5 &&  pid->neuronVal()<3.5 )         //the particle is pion if 1.5<pid->neuronVal()<3.5      if( pid->neuronVal()>3.5 &&  pid->neuronVal()<4.5 )         //the particle is kaon if 3.5<pid->neuronVal()<4.5        if( pid->neuronVal()>4.5 &&  pid->neuronVal()<5.5 )         //the particle is proton if 4.5<pid->neuronVal()<5.5        if( pid->neuronVal()>6)       //the particle is muon if pid->neuronVal()>6

Chinese Physics C 1,32 (2008)

/afs/ihep.ac.cn/bes3/***/6.1.0/Analysis/BParticleAlg/**/share/

Read MC Truth Information

Requirement File: use McTruth McTruth-* Event

***.cxx file: #include "McTruth/McParticle.h"

//*** MCMADE information SmartDataPtr<Event::McParticleCol> mcParticleCol(eventSvc(), "/Event/MC/McParticleCol"); if (!mcParticleCol){ log<< MSG::FATAL << "Could not find McParticle" << endreq; return( StatusCode::FAILURE); } Event::McParticleCol::iterator iiter_mc = mcParticleCol->begin(); for(;iiter_mc != mcParticleCol->end(); iiter_mc++){ double px=(*iiter_mc)->initialFourMomentum().px()/1000.; double py=(*iiter_mc)->initialFourMomentum().py()/1000.; double pz=(*iiter_mc)->initialFourMomentum().pz()/1000.; double ee=(*iiter_mc)->initialFourMomentum().e()/1000.; double mm=sqrt(abs(ee*ee-px*px-py*py-pz*pz));

pdg=(*iiter_mc)->particleProperty();

if(pdg==441) //etac { p4etac.setPx(px); p4etac.setPy(py); p4etac.setPz(pz); p4etac.setE(ee); }

} //for

Most interesting?

Prices measurement of mass of ηc can provide information on the hyperfine (spin-spin) splitting of ηc and J/Ψ

C.L. = 0.006

C.L. = 0.001Br(J/c) is off too!

Known problems in BOSS6.3.1

Using J/Ψ→γηc, ηc→ηπ+π- as example to check. Why study ηc decays?

Our Goal:Measure the inclusive properties of the c with > 10x improved precision

1)Mc: ± 1.2 MeV ± 0.1 MeV)tot: ± 3.5 MeV ± 0.5 MeV)BrJ/c: ± .36% ± 0.05%

Priority order The expected

statistical error of c mass measurement is about 0.05 MeV at BESIII.

General selection:

Good charged tracks select: Vr < 1.0cm, |Vz|<5.0cm, |cosθ|<0.93

Good photon select: Eraw >= 40MeV,

Ntrk=2 and ΣQ=0

PID: pi: Prob(K) <Prob(pi)

4C Kinematic Fit, looping over all good photon candidates, select the combination with the minimum χ2.

20||or 20|| cncn

● <χ2>≠4. So the χ2 value problem not solved yet in Boss 6.3.1

● We do not add the χ2 cut, the default value is 500.

χ distributions are not correct, so the PID efficiency is not reliable.

But in BOSS6.3.2

Inclusive MC

2.0M

PYTHIA

MC

Cut on 2 is not used!

Chi distribution in PID Boss6.3.2

χ distributions are correct!

MeV2.16 MeV5.25

Boss6.3.2 Boss6.3.1

BESIII Physics Yellow Book